LSTM-ED for Anomaly Detection in Time Series Data¶

In [ ]:
import time
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import tensorflow as tf

from dataset import *
from plots import *
from metrics import *
from models_funtions import *

# Set style for matplotlib
plt.style.use("Solarize_Light2")

import plotly.io as pio
pio.renderers.default = "notebook_connected"
In [ ]:
# Path to the root directory of the dataset
ROOTDIR_DATASET_NORMAL =  '../dataset/normal'
ROOTDIR_DATASET_ANOMALY = '../dataset/collisions'

# TF_ENABLE_ONEDNN_OPTS=0 means that the model will not use the oneDNN library for optimization

import os
os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0'

Variours parameters¶

In [ ]:
#freq = '1.0'
#freq = '0.1'
freq = '0.01'
#freq = '0.005'

file_name_normal = "_20220811_rbtc_"
file_name_collisions = "_collision_20220811_rbtc_"

recording_normal = [0, 2, 3, 4]
recording_collisions = [1, 5]

freq_str = freq.replace(".", "_")
features_folder_normal = f"./features/normal{freq_str}/"
features_folder_collisions = f"./features/collisions{freq_str}/"

Data¶

In [ ]:
df_features_normal, df_normal_raw, _ = get_dataframes(ROOTDIR_DATASET_NORMAL, file_name_normal, recording_normal, freq, f"{features_folder_normal}")
df_features_collisions, df_collisions_raw, df_collisions_raw_action = get_dataframes(ROOTDIR_DATASET_ANOMALY, file_name_collisions, recording_collisions, freq, f"{features_folder_collisions}1_5/")
df_features_collisions_1, df_collisions_raw_1, df_collisions_raw_action_1 = get_dataframes(ROOTDIR_DATASET_ANOMALY, file_name_collisions, [1], freq, f"{features_folder_collisions}1/")
df_features_collisions_5, df_collisions_raw_5, df_collisions_raw_action_5 = get_dataframes(ROOTDIR_DATASET_ANOMALY, file_name_collisions, [5], freq, f"{features_folder_collisions}5/")
Loading data.
Found 31 different actions.
Loading data done.

Loading features from file.
--- 0.03563117980957031 seconds ---
Loading data.
Found 31 different actions.
Loading data done.

Loading features from file.
--- 0.022705078125 seconds ---
Loading data.
Found 31 different actions.
Loading data done.

Loading features from file.
--- 0.01462244987487793 seconds ---
Loading data.
Found 31 different actions.
Loading data done.

Loading features from file.
--- 0.013123512268066406 seconds ---
In [ ]:
X_train, y_train, X_test, y_test, df_test = get_train_test_data(df_features_normal, df_features_collisions, full_normal=True)
X_train_1, y_train_1, X_test_1, y_test_1, df_test_1 = get_train_test_data(df_features_normal, df_features_collisions_1, full_normal=True)
X_train_5, y_train_5, X_test_5, y_test_5, df_test_5 = get_train_test_data(df_features_normal, df_features_collisions_5, full_normal=True)
c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\.venv\Lib\site-packages\sklearn\base.py:493: UserWarning:

X does not have valid feature names, but VarianceThreshold was fitted with feature names

c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\.venv\Lib\site-packages\sklearn\base.py:493: UserWarning:

X does not have valid feature names, but VarianceThreshold was fitted with feature names

c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\.venv\Lib\site-packages\sklearn\base.py:493: UserWarning:

X does not have valid feature names, but VarianceThreshold was fitted with feature names

Collisions¶

In [ ]:
collisions_rec1, collisions_init1 = get_collisions('1', ROOTDIR_DATASET_ANOMALY)
collisions_rec5, collisions_init5 = get_collisions('5', ROOTDIR_DATASET_ANOMALY)

# Merge the collisions of the two recordings in one dataframe
collisions_rec = pd.concat([collisions_rec1, collisions_rec5])
collisions_init = pd.concat([collisions_init1, collisions_init5])
In [ ]:
collisions_zones, y_collisions = get_collisions_zones_and_labels(collisions_rec, collisions_init, df_features_collisions)
collisions_zones_1, y_collisions_1 = get_collisions_zones_and_labels(collisions_rec1, collisions_init1, df_features_collisions_1)
collisions_zones_5, y_collisions_5 = get_collisions_zones_and_labels(collisions_rec5, collisions_init5, df_features_collisions_5)

LSTM-ED for Anomaly Detection in Time Series Data¶

In [ ]:
from algorithms.lstm_enc_dec_axl import LSTMED

classifier = LSTMED(
    name='LSTM-ED',
    num_epochs=50,
    batch_size=64,
    lr=1e-3,
    hidden_size=64,
    sequence_length=100,
    train_gaussian_percentage=0.30,
    n_layers=(2, 2),
    use_bias=(True, True),
    dropout=(0.1, 0.1),
    seed=42,
    gpu=None,              # Set to None for CPU, or specify GPU index if available
    details=True
)
# Train the LSTM on normal data
classifier.fit(X_train)
print("LSTM-ED training completed.")
  0%|          | 0/50 [00:00<?, ?it/s]c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\.venv\Lib\site-packages\torch\nn\_reduction.py:42: UserWarning:

size_average and reduce args will be deprecated, please use reduction='sum' instead.

100%|██████████| 50/50 [01:38<00:00,  1.97s/it]
c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\.venv\Lib\site-packages\torch\nn\_reduction.py:42: UserWarning:

size_average and reduce args will be deprecated, please use reduction='none' instead.

LSTM-ED training completed.

Predictions¶

In [ ]:
df_test = get_statistics(X_test, y_collisions, classifier, df_test, freq, threshold_type="mad")
df_test_1 = get_statistics(X_test_1, y_collisions_1, classifier, df_test_1, freq, threshold_type="mad")
df_test_5 = get_statistics(X_test_5, y_collisions_5, classifier, df_test_5, freq, threshold_type="mad")
Anomaly prediction completed.
Number of anomalies detected: 3 with threshold 37860.8129479301, std
Number of anomalies detected: 119 with threshold 162.11054691809508, mad
Number of anomalies detected: 16 with threshold 571.5824529741639, percentile
Number of anomalies detected: 8 with threshold 746.712019359903, IQR
Number of anomalies detected: 306 with threshold 0.0, zero

choosen threshold type: mad, with value: 162.1105
F1 Score: 0.9375
Accuracy: 0.9542
Precision: 0.8824
Recall: 1.0000
              precision    recall  f1-score   support

           0       1.00      0.93      0.96       201
           1       0.88      1.00      0.94       105

    accuracy                           0.95       306
   macro avg       0.94      0.97      0.95       306
weighted avg       0.96      0.95      0.95       306

ROC AUC Score: 0.9743
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
Anomalies detected: 119
Best threshold: 165.5587 | F1 Score: 0.9375 | Precision: 0.8824 | Recall: 1.0000
Anomalies detected with best threshold: 119

	-------------------------------------------------------------------------------------

c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\src\models_funtions.py:67: RuntimeWarning:

invalid value encountered in divide

c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\.venv\Lib\site-packages\torch\nn\_reduction.py:42: UserWarning:

size_average and reduce args will be deprecated, please use reduction='none' instead.

Anomaly prediction completed.
Number of anomalies detected: 1 with threshold 29341.05882837139, std
Number of anomalies detected: 45 with threshold 136.12575708314188, mad
Number of anomalies detected: 9 with threshold 483.22738789366224, percentile
Number of anomalies detected: 22 with threshold 271.07064919633075, IQR
Number of anomalies detected: 164 with threshold 0.0, zero

choosen threshold type: mad, with value: 136.1258
F1 Score: 0.8750
Accuracy: 0.9390
Precision: 0.7778
Recall: 1.0000
              precision    recall  f1-score   support

           0       1.00      0.92      0.96       129
           1       0.78      1.00      0.88        35

    accuracy                           0.94       164
   macro avg       0.89      0.96      0.92       164
weighted avg       0.95      0.94      0.94       164

ROC AUC Score: 0.9818
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
Anomalies detected: 45
Best threshold: 160.2134 | F1 Score: 0.9091 | Precision: 0.8333 | Recall: 1.0000
Anomalies detected with best threshold: 42

	-------------------------------------------------------------------------------------

c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\src\models_funtions.py:67: RuntimeWarning:

invalid value encountered in divide

c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\.venv\Lib\site-packages\torch\nn\_reduction.py:42: UserWarning:

size_average and reduce args will be deprecated, please use reduction='none' instead.

Anomaly prediction completed.
Number of anomalies detected: 2 with threshold 46117.9134284106, std
Number of anomalies detected: 10 with threshold 561.8715677574588, mad
Number of anomalies detected: 8 with threshold 602.4740841487306, percentile
Number of anomalies detected: 3 with threshold 931.1136584338797, IQR
Number of anomalies detected: 141 with threshold 0.0, zero

choosen threshold type: mad, with value: 561.8716
F1 Score: 0.2121
Accuracy: 0.6312
Precision: 0.7000
Recall: 0.1250
              precision    recall  f1-score   support

           0       0.63      0.96      0.76        85
           1       0.70      0.12      0.21        56

    accuracy                           0.63       141
   macro avg       0.66      0.54      0.49       141
weighted avg       0.66      0.63      0.54       141

ROC AUC Score: 0.9223
No description has been provided for this image
No description has been provided for this image
No description has been provided for this image
Anomalies detected: 10
Best threshold: 338.7640 | F1 Score: 0.9076 | Precision: 0.8571 | Recall: 0.9643
Anomalies detected with best threshold: 63

	-------------------------------------------------------------------------------------

c:\Users\VG User\Documents\GitHub\MLinAPP-FP01-14\src\models_funtions.py:67: RuntimeWarning:

invalid value encountered in divide

In [ ]:
plot_anomalies_true_and_predicted(df_collisions_raw, df_collisions_raw_action, collisions_zones, df_test, title="Collisions zones vs predicted zones for both recordings")
In [ ]:
plot_anomalies_true_and_predicted(df_collisions_raw_1, df_collisions_raw_action_1, collisions_zones_1, df_test_1, title="Collisions zones vs predicted zones for recording 1")
In [ ]:
plot_anomalies_true_and_predicted(df_collisions_raw_5, df_collisions_raw_action_5, collisions_zones_5, df_test_5, title="Collisions zones vs predicted zones for recording 5")